Eric has done these until now, but I now find myself with a few small things that fit well into such a post.
- Older SIGGRAPH Courses often have great material in them, but are tough to track down. This website has a bunch of links to course notes from 1999 to 2007.
- The SIGGRAPH Education Committee has a page with links to a few even older courses, going back to 1996. The “Pixel Cinematography” course from 1996 looks especially interesting.
- Fabian “ryg” Giesen is doing a great series of posts (as yet unfinished) on his blog, which take the reader on A trip through the graphics pipeline. He recently started reposting a slightly cleaned up version of the series on AltDevBlogADay.
- A variant on a previously published paper (video here), Deferred Screen-Space Directional Occlusion by Yuriy O’Donnell has increased performance and plugs relatively easily into deferred shading pipelines.
- Emil “Humus” Persson has recently released a demo of his Geometry Buffer Anti-Aliasing technique, which he will also be presenting at an upcoming SIGGRAPH course.
- I’ve long been interested in the problem of filtering normals in a way that correctly accounts for surface appearance; we also discuss this in Section 7.8.1 of Real-Time Rendering. Stephen Hill has kicked off his new blog with an excellent post summarizing various solutions to the problem, including his own solution as well as a WebGL demo. The comments to the post are also well worth reading; a lively discussion has developed, with Brian Karis of Human Head Studios describing the solution used on the upcoming game Prey 2.
- One of the techniques discussed in the aforementioned post was LEAN mapping and its lighter-weight variant CLEAN mapping. Inspired by that post, Marc Olano (first author on the LEAN mapping paper) has posted some of his own thoughts on those techniques.
The trip through the graphics pipeline has a lot of detail, though I wish he had spent time explaining some of the intermediate steps and why they’re there a bit more. Still, great information in one place that I’ve never seen elsewhere (at least not in a fairly comprehensible format).
Marc Olano just posted a follow-up on the LEAN/CLEAN mapping stuff: http://gaim.umbc.edu/2011/07/26/on-error/